« get me outta code hell

hsmusic-wiki - HSMusic - static wiki software cataloguing collaborative creation
about summary refs log tree commit diff
AgeCommit message (Collapse)Author
2021-04-19transform replacer arguments tf-parser(quasar) nebula
2021-04-19trim whitespace surrounding first/last text nodes(quasar) nebula
2021-04-19move textOnly tag test before advancing past "[["(quasar) nebula
This makes syntax error messages a little nicer, i.e. hello there [[album:[[lol]]]] --------------------^ instead of hello there [[album:[[lol]]]] ----------------------^
2021-04-19fix incorrect iEnd values on text nodes(quasar) nebula
The iEnd value (which marks the position a node's length ends at in input text) is only used for tag nodes at the moment, 8ut it's still nice to have the right values everywhere, in case we do use it for something else later (and to just have code that functions like we want, lol).
2021-04-16connect parser to updated replacer & rest of code!(quasar) nebula
2021-04-16show line and column position in parse errors(quasar) nebula
2021-04-16match hash/replacerValue only without preceding WS(quasar) nebula
2021-04-16add end index to node info(quasar) nebula
2021-04-16add hash to tag syntax(quasar) nebula
2021-04-16fix final text not being added as node(quasar) nebula
2021-04-15simple replaceKey/Value syntax (cur: 110k/sec)(quasar) nebula
Not actually a speed-up, 8ut this is easier to read. Learn it from me: with no other 8enefits, prefer what's legi8le, not what's cool! At least in projects where the whole point isn't 8eing cool, anyway.
2021-04-15remove unused original code(quasar) nebula
2021-04-15use regexes, avoid parse by letter (cur: 110k/sec)(quasar) nebula
2021-04-15don't global-hoist speedtest definition(quasar) nebula
2021-04-15run 5s of tests instead of 1 (cur: 87k/sec)(quasar) nebula
Why does this improve efficiency? I do not know. It's entirely possi8le the earlier estim8tions were just off, and this is more accur8te.
2021-04-15parser error throwing/handling bugfixes(quasar) nebula
2021-04-15don't use makeNode (cur: 84k/sec)(quasar) nebula
This actually might 8e 85k/sec 8ut that feels, uh, counter-intutive? Pretty sure it was just getting optimized out, anyway.
2021-04-15remove void parseOneTextNode logic (cur: 84k/sec)(quasar) nebula
If textOnly is passed, parseNodes will only ever return a single text node, so the other paths were never getting reached. This is only cleanup, though - no significant performance improvement.
2021-04-15make data in text nodes string (cur: 84k/sec)(quasar) nebula
This doesn't actually impact the performance at all, 8ut it's going to 8e nicer to work with later.
2021-04-15remove objs from fn signatures (cur: 84k/sec)(quasar) nebula
Hoo 8*tch, that's a 8ig improvement! This gets rid of the last spread/destructure, a straggler in parseOneTextNode, and makes the only non-primitive passed to either itself or parseNodes the array of tags to parse, cutting away a 8unch of o8jects which were only getting used the one time.
2021-04-15store node data in its own property (cur: 51k/sec)(quasar) nebula
Another place to avoid destructuring (actually the spread operator, i.e. effectively an Object.assign call).
2021-04-15simplify once-used pushNode call (cur: 48.5k/sec)(quasar) nebula
The idea here is to get rid of unnecessary destructuring of the arguments array.
2021-04-15don't recreate reused functions (cur: 48k/sec)(quasar) nebula
This commit 8est viewed with whitespace changes hidden!
2021-04-15no unnecessary objects in returns (cur: 46k/sec)(quasar) nebula
2021-04-15literal matching as plain strings (cur: 40k/sec)(quasar) nebula
2021-04-15new parser progress so far (cur: 15k/sec)(quasar) nebula
not implemented into the site (so just process.exit()), includes a couple examples and speedtest (not optimized at all yet)
2021-04-10new Albums - by Date Added to Wiki listing(quasar) nebula
2021-04-10hide unreleased tracks from by-duration listing(quasar) nebula
For........ reasons. ::::) 8ut also, pro8a8ly makes sense to do this anyway.
2021-04-09New Additions homepage row & date added to wiki(quasar) nebula
2021-04-09disable end-of-page banner on track pages(quasar) nebula
2021-04-09hide info cards behind localStorage flag for now(quasar) nebula
2021-04-06only animate info card hiding if it's shown(quasar) nebula
This fixes a glitch where it'd trigger the hide animation just by passing the cursor over the link for even an instant. Not good!
2021-04-06show cover artists in info card(quasar) nebula
2021-04-06auto-height image banner(quasar) nebula
this isn't a 100% perfect fix since loaded, upscaled banners will have a *slightly* different final height than the lazy load (about 3px), but it's relatively negligible. preferably, refined banners would be the target 1100x200 aspect ratio in the first place.
2021-04-06fix padding from info card at bottom of page(quasar) nebula
2021-04-06banner style tweaks(quasar) nebula
2021-04-06fix banner waiting to load messing up layout(quasar) nebula
2021-04-06fix broken linkPathname-style links(quasar) nebula
2021-04-06fix colorless album groups in sidebar (thanks FF!)(quasar) nebula
2021-04-05i fixed the bug :)(quasar) nebula
2021-04-05basic artist list in info cards(quasar) nebula
2021-04-05now That is a typo(quasar) nebula
2021-04-05use or mock ListFormat api on client(quasar) nebula
2021-04-05reveal strings on info card art(quasar) nebula
2021-04-05info card thumbnail & image link/color niceness(quasar) nebula
2021-04-05info card dimension tweaks(quasar) nebula
2021-04-04cover art in info card(quasar) nebula
2021-04-04fix Tags - by Uses listing directory lol(quasar) nebula
2021-04-04more url system fixes(quasar) nebula
slowly approaching something not terrible here
2021-04-04info card animations(quasar) nebula